The combination of poor diet and sedentary lifestyles is second only to smoking in causing preventable health problems and deaths in the United States[a].¹ The consequences go well beyond health, impacting quality of life, productivity, and the general economy. These effects can be avoided through more active lifestyles, combined with good eating habits.

Until this survey was conducted, only a little more than anecdotal information was known about eating and physical activity habits within Marion County. There were many outcomes indicating unhealthy habits, including: high rates of diabetes and other chronic diseases; high state-level rates of obesity and obesity-related Medicaid expenditures[b]; and county-wide obesity rate estimates that exceed the rest of Indiana. However, these estimates were based on too few people to adequately assess which groups in Marion County have the greatest problems. There was also little information about the habits, attitudes, and factors contributing to obesity among our 870,000 residents.

A generous grant from the Richard M. Fairbanks Foundation made it possible for the Marion County Health Department (MCHD) to conduct the first comprehensive adult obesity needs assessment telephone survey. Between February and June, 2005, telephone interviews of 4,784 adults in Marion County were conducted to establish information about body mass index (BMI), physical activity behaviors, food intake, eating patterns, and other related factors. The result is a very rich data set, sufficient to assess physical activity, nutritional status, obesity, in many population subgroups, as well as for the county as a whole. Highlights from the analyses are included in this report.

General Results

Survey results indicate that one-quarter of Marion County adults were obese (25.6%), and 60 percent were either overweight or obese. Obesity was common in all demographic groups, especially among Black females, the middle-aged, persons with lower incomes, and those with chronic health problems.

Eating habits were poor, though there are indications that many people would like to improve their dietary habits. Few Marion County adults, particularly obese adults, ate as many fruits and vegetables as are recommended, and most people did not pay attention to the serving size on food labels. However, most people paid at least some attention to nutrition labels, and looked for nutrition information at restaurants. Among those who did not practice healthful eating, most intended to improve their eating.

While most of the population was at least somewhat physically active, a large portion was not. Most relatively inactive people would like to increase their activity levels. Many people were physically active in their jobs, and most people had safe, convenient, affordable places to exercise. Between one-third and one-half of adults walked for exercise at least a few times per week. But four out-of-ten obese adults and three out-of-ten other adults did not achieve the United State’s Centers for Disease Control and Prevention (CDC) recommended level of physical activity. Almost half of those adults intended to increase their level of activity within 30 days.

Poverty was clearly related to nutrition and physical activity barriers. Three out-of-ten adults with low incomes had poor access to fruits and vegetables. Four out-of-ten reported that they had no safe, affordable, convenient place to exercise.

Body Weight

In Marion County, one out-of- four adults (25.6%) who responded to the survey was obese. Six out-of-ten adults were overweight or obese (60.7%). These percentages are similar to state and national statistics. The data were analyzed by race, gender, age, income, education, and area of residence within the county. The greatest differences in BMI[c] occurred in the race by gender analysis, with Black females having notably higher rates of obesity than any other demographic group.

· Females, by race: Eight out-of-twenty (39%) Black females were obese, and over two-thirds (69%) were obese or overweight. In comparison, less than five out-of-twenty (23%) White or Latino females were obese, and ten out-of-twenty (49% and 50%, respectively) were obese or overweight

· Males, by race: There was less difference in obesity by race among males. Black males had a few percentage points more obesity (28%) than White (24%) or Latino men (19%) (Figure 5).

· Age: The youngest adults (less than 25 years of age) and oldest (over 75-years-old) had the least obesity (16%), while middle-aged persons (45 to 64 years old) had the most (31%) (Figure 6).

· Education: Level of education had a moderate influence, with college graduates having about seven percentage points less obesity than those with less education (21% versus about 28%).

· Income: The prevalence of obesity also declined as income increased, from 34 percent for the lowest income adults to 23 percent for adults with higher income.

· Geographic area: Obesity prevalence also varied by area within the county, with prevalence in the low 30-percent range occurring in a band that ran roughly east-west across the county, just north of the middle of the county. Most other areas had prevalence in the mid to-low-20 percent range.

· Weight change: Females were more likely than males to consider themselves overweight, even among those in the Normal or Underweight BMI category. Half of overweight males and one out-of-seven obese males considered themselves at “about the right weight”. About one third of overweight and obese males and females had tried to lose weight in the prior year.

· Health effects: Increased weight was clearly associated with increased chronic disease. Chronic disease was almost twice as common among those in the Obese BMI category compared to those in the Normal or Underweight BMI category.

· Advice from clinicians: One-third of obese adults and two-thirds of overweight adults reported that a doctor had not told them that they were overweight. Many obese adults (over 40%) and most adults who are not obese (50% to 75%) had not received nutrition or physical activity advice from their doctors.

Diet and Nutrition

· Healthy eating: Most adults reported eating fewer fruits and vegetables than are recommended by the CDC, and more red meat than many nutrition experts advise. Overweight, and especially obese adults ate more red meat and fewer fruits and vegetables than did adults in the Normal or Underweight BMI category. Three out-of-four adults ate less fruit than is recommended, and only one-in-twenty adults ate as many vegetables as are recommended. However, most obese adults were planning to change their eating habits to eat a more healthy diet.

· Nutrition labels and serving size: While two-thirds of adults read nutrition labels at least some of the time when buying groceries (Figure 20), only half of adults looked at the serving size on the label at least some of the time in deciding how much to eat (Figure 21). About half also looked for nutrition information at restaurants, at least some of the time.

· Access to fruits and vegetables: Most adults had convenient access to a store selling fresh fruits and vegetables, though 15 percent of adults with incomes below the poverty level[d] did not. The produce at these stores was affordable for most people, though it was not affordable for the 10 percent of adults whose household income was in the lower third of Marion County household incomes.[e] Combining these factors, 13 percent of Marion County adults, including 31 percent of adults living in poverty, did not have convenient access to affordable fruits and vegetables.

Physical Activity

· Amount of physical activity: As might be expected, adults with lower BMI values were more physically active and more frequently met the CDC’s minimum recommended activity level. The percent that met the CDC’s minimum recommended level of physical activity tended to decrease with age. Except among adults less than age 25, obese adults met recommended physical activity levels comparable to non-obese adults who were twenty to thirty years older.

· Readiness to change: Almost half of those who did not meet the CDC’s minimum recommended activity level indicated that they intended to increase their physical activity level in the next 30 days. Adults in the Obese BMI category were most likely to report that they intend to increase their physical activity, but they were less likely to have maintained an increase for longer than six months.

· Physical activity at work: Employment offered many people opportunities to be physically active. Three out-of-five employed adults had jobs that kept them physically active. About 30 percent of adults in each BMI category had access to a place to exercise at work, and a similar proportion had employer incentives to exercise.

· Physical activity in the community: Eight-of-ten respondents reported having a safe, convenient and affordable place to exercise in their communities. No two race and gender groups differed by more than ten percent in their access to such places. Access declined however as household income decreased, especially for those in the Overweight and Obese BMI categories. Less than 60 percent of overweight and obese adults in households with incomes below the federal poverty level had safe, convenient and affordable places to exercise.

· Safety concerns about exercising in their community: Among adults reporting no safe, convenient and affordable place to exercise, only one-in-twenty) had safety concerns about exercising in their community. Convenience and/or affordability may have been barriers for the other 19. As with other barriers, the safety barrier was greatest for obese and overweight adults with very low income, who were two-to-six times more likely to have safety concerns as any other group.

· Walking in neighborhood: Between one-third and one-half of adults walked at least weekly in their neighborhood for exercise. Those in the higher BMI categories were least likely to walk in their neighborhoods. Respondents with household incomes below the federal poverty level were most likely to walk in their neighborhood every day. This may reflect a lack of personal transportation for this lowest income group. Among race and gender groups, Black females were the most likely to never walk in their neighborhood for exercise (36%), while Black males were most likely to walk daily in their neighborhood for exercise (27%).

Introduction

Figure 1: Adult Obesity Prevalence in Indiana and the United States, 1990-2005

Source: CDC BRFSS http://www.cdc.gov/brfss/

In 1990, about one eighth (11.6%) of adults in the United States were obese. In 2005, about one-quarter of adults was obese. The prevalence of obesity had doubled in just 15 years. Obesity in Indiana has consistently been more common than in the rest of the country, by about three percentage points.

Obesity and Health

The combined effect of poor diet, lack of physical activity, and excessive body fat is second only to tobacco in causing premature death in the United States.¹ These modifiable risk factors contribute to many chronic diseases and conditions, increase health care costs, decrease productivity and quality of life, and result in disability, suffering, and other far-reaching economic and societal costs.

Overweight and Obesity

· Two-thirds of all diabetes, 20 percent of cancer in women, and 14 percent of cancer in men can be attributed to being overweight or obese ⁶.

· Fatty tissue produces a variety of hormones, affecting wide-ranging body processes from tumor growth, to insulin response, to blood lipids, to blood vessel tone and blood pressure, to inflammation and blood clotting effects.^{2, 7}

· Obese women have nearly four times the risk of knee osteoarthritis (OA) as non-obese women; obese men have nearly five times the risk of OA as non-obese men, and risk of OA is up to 10 times higher for those at the highest quintile of weight (the heaviest 20% of men and women).^{8, 9}

· Women with a Body Mass Index (BMI[f]) of 29 or over are 3 times more likely to have coronary artery disease than women with a BMI of 21 or less.¹⁰

· There is a 10 percent increase in risk of a coronary event, e.g. myocardial infarction (heart attack), in men for each unit increase in BMI above 22.¹⁰

· High blood pressure prevalence doubles as BMI increases from less than 25 to greater than 30.¹⁰

· Stroke risk doubles as BMI increases from 23 to 30.¹⁰

· Type 2 diabetes risk doubles to triples with increased BMI from 22 to 30. Men and women with a BMI of 35 or more are 20 times more likely to develop diabetes than those with a BMI in the normal range.^{3, 10}

· Men and women who are overweight but not obese (BMI 25-29.9) have a 1.5 to2 times greater risk of developing high cholesterol, high blood pressure, colon cancer, and gallstones than their peers with BMI less than 25.³

Overweight and obesity are also associated with:

· Age-related macular degeneration

· Asthma

· Bladder control problems

· Cancers: colon, rectum, breast, uterus, prostate, kidney, liver, pancreas, esophagus

· Congestive heart failure

· Depression

· Fatty liver disease

· Gallbladder disease

· Gout

· Infertility

· Menstrual irregularities

· Musculoskeletal disorders

· Pregnancy complications

· Sleep apnea/respiratory problems

Together, cardiovascular disease, cancer, and diabetes account for approximately two-thirds of all deaths in the United States. The direct and indirect costs for these diseases are estimated to be about $700 billion each year², $61 billion for direct medical costs and $56 billion for indirect costs of absenteeism and lost productivity.¹¹

Physical Activity

Physical activity reduces the risk for many chronic diseases and conditions through its impact on weight management. In addition, physical activity provides health benefits independent of weight control. Some of these benefits are the result of the effects of regular physical activity on hormones, including estrogen, androgens, and insulin. Other benefits include improving the health of the cardio-respiratory system, decreasing the time it takes food to pass through the intestines, and preventing type 2 diabetes. Research continues to reveal additional positive effects.² Health benefits increase with the amount of time an individual is active and with the intensity of physical activity, up to a point.

Specific examples of the proven effects of regular, moderate physical activity (walking 3 to 4.5 miles-per-hour for 30 minutes or longer, five or more days per week) include:

· Helps control weight.

· Helps control appetite.

· Helps control high blood pressure.

· Reduces risk for type 2 diabetes.

· Helps control type 2 diabetes.

· Reduces risk of coronary artery disease and second heart attacks.

· Reduces arthritis pain and disability.

· Reduces risk for osteoporosis and falls.

· Helps the elderly maintain independence.

· Reduces risk of stroke.

· Reduces risk of colon cancer.

· Reduces symptoms of depression, stress, and anxiety.

· Improves memory and brain function in young and old.

· Maintains muscle mass.

Supervised physical activity can improve heart function in many patients with heart failure and improve lower extremity circulation in patients with peripheral arterial disease.^{12, 13}

Diet and Nutrition

Dietary choices also affect health, both directly, and indirectly through their impact on weight. Decreased risk of a variety of chronic diseases are associated with diets rich in fruits and vegetables, whole grain foods, and legumes, and low in red meat, other saturated animal fats, dairy fat, and added sugars ¹⁴. Attempts to isolate specific nutrients found in fruits and vegetables and deliver them as a pill or other form or supplement generally have been unsuccessful in providing the same health benefits, and have proven harmful in some cases. These experiences demonstrate that whole foods are complex delivery systems, and they remain the best sources for these beneficial nutrients.

Over the last two decades in the United States, there has been a decrease in fruit and vegetable consumption and an increase in dietary fat intake. In other words, people are increasingly eating foods that are associated with greater risks of major chronic diseases. Americans are consuming more calories per day, with the largest percentage of the increase in calories since the 1980s coming from an increase in consumption of refined grains and foods high in added sugar.^{2, 14}

Given the harmful effects of poor diet, lack of physical activity, overweight and obesity, and the proven benefits of healthier lifestyles, it is critical that we understand the behaviors and concerns of Marion County citizens regarding diet, physical activity, and weight. With this information, we may identify and address those factors that best promote healthy lives.

Bibliography

1. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA Mar 10 2004;291(10):1238-1245.

2. Eyre H, Kahn R, Robertson RM. Preventing cancer, cardiovascular disease, and diabetes: a common agenda for the American Cancer Society, the American Diabetes Association, and the American Heart Association. CA Cancer J Clin. Jul-Aug 2004;54(4):190-207.

3. Field AE, Coakley EH, Must A, et al. Impact of overweight on the risk of developing common chronic diseases during a 10-year period. Arch Intern Med. Jul 9 2001;161(13):1581-1586.

4. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. Feb 7 2002;346(6):393-403.

5. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. May 3 2001;344(18):1343-1350.

6. Oster G, Thompson D, Edelsberg J, Bird AP, Colditz GA. Lifetime health and economic benefits of weight loss among obese persons. Am J Public Health. Oct 1999;89(10):1536-1542.

7. Wisse BE. The inflammatory syndrome: the role of adipose tissue cytokines in metabolic disorders linked to obesity. J Am Soc Nephrol. Nov 2004;15(11):2792-2800.

8. Anderson JJ, Felson DT. Factors associated with osteoarthritis of the knee in the first national Health and Nutrition Examination Survey (HANES I). Evidence for an association with overweight, race, and physical demands of work. Am J Epidemiol. Jul 1988;128(1):179-189.

9. Felson DT. Weight and osteoarthritis. J Rheumatol Suppl. Feb 1995;43:7-9.

10. Troiano RP. PA-01-017: Physical Activity and Obesity Across Chronic Diseases. Department of Health and Human Services. Available at: http://grants.nih.gov/grants/guide/pa-files/PA-01-017.html. Accessed October 30, 2005.

11. Sturm R. The effects of obesity, smoking, and drinking on medical problems and costs. Health Aff (Millwood). Mar-Apr 2002;21(2):245-253.

12. Division of Nutrition and Physical Activity NCfCDPaHP. Physical Activity for Everyone: Home | DNPA | CDC. CDC. 09/29/2005. Available at: http://www.cdc.gov/nccdphp/dnpa/physical/index.htm. Accessed October 30, 2005.

13. Thompson PD, Buchner D, Pina IL, et al. Exercise and physical activity in the prevention and treatment of atherosclerotic cardiovascular disease: a statement from the Council on Clinical Cardiology (Subcommittee on Exercise, Rehabilitation, and Prevention) and the Council on Nutrition, Physical Activity, and Metabolism (Subcommittee on Physical Activity). Circulation. Jun 24 2003;107(24):3109-3116.

14. Byers T, Nestle M, McTiernan A, et al. American Cancer Society guidelines on nutrition and physical activity for cancer prevention: Reducing the risk of cancer with healthy food choices and physical activity. CA Cancer J Clin. Mar-Apr 2002;52(2):92-119.

Methods

Purpose

The purpose of the Marion County Adult Obesity Needs Assessment Telephone Survey was to gather information about citizens’ body mass index (BMI) values, physical activity behaviors, food intake, eating patterns, and other related factors for various population subgroups, and population as a whole in Marion County. This information will be used throughout the community to develop initiatives to reduce obesity and overweight, and serve as a baseline, against which future survey results can be compared, to assess changes in obesity and obesity-related factors.

Survey Methods

Background

The 2005 Marion County Adult Obesity Needs Assessment Telephone Survey collected baseline data on a wide range of factors associated with overweight and obesity among adult citizens of Marion County, Indiana. The Indiana University Public Opinion Laboratory conducted the telephone survey using trained telephone survey interviewers and the Computer Assisted Telephone Interview (CATI) software system.

Instrument

The items on the survey instrument and the survey methodology were developed by a working committee comprised of MCHD staff, Indiana University Bowen Research Center staff and other key individuals, with the input of a community advisory group. The investigators designed the questionnaire. Most of the items selected were from other validated and publicly available instruments, primarily the CDC’s Behavioral Risk Factor Surveillance System. A Spanish language version of the instrument was also developed for respondents preferring to answer questions in that language. Health status, respondent demographics, opportunities for physical activity, amount of physical activity, food availability and consumption, eating habits, advice from the doctor, tobacco use, mental health, education, and income were measured. The instrument also contained questions to verify county of residence and screen for age. The instrument was pre-tested and revisions were made prior to its use in the study.

Subjects

The target population included English- and Spanish-speaking adults (18 or older) residing in Marion County. Targeted individuals were classified as “unable to contact” following interviewer completion of seven attempts at different times of the day and different days of the week. The questionnaire was completed by 34.1 percent of the potential respondents who were successfully contacted. All together, 4,784 Marion County residents were surveyed during the six months between February and July 2005. Approximately 800 telephone surveys were completed each month, to assess seasonal variations in responses.. To ensure appropriate representation in the results, over-sampling was used to obtain at least 400 respondents from each gender among non-Latino Caucasians and among African Americans, and at least 200 respondents from both Latino males and Latino females. The survey responses were weighted to ensure that the results reflected the actual distribution of the county population by race, ethnicity, and gender Pregnant women were excluded from the survey, as pregnancy has a strong impact on weight, physical activity, and diet, and requires a separate study.

Institutional Review Board and Human Subjects

All investigators were required to have an Indiana University Purdue University Indianapolis Protection of Human Subjects in Research certification. The Indiana University Institutional Review Board approved the study protocol [IRB Study Number 0502-56] in February 2005, before the survey began.

Demographics

Table 1 below shows the number of respondents by gender, race, age, education, and household income. It also includes the weighted percent of respondents in each of these categories. Most respondents were women, as is often the case in telephone surveys. With a total of 4,784 respondents, there were a sufficient number of respondents in many subgroups for reliable estimates to be produced about those subgroups, even after stratification for such variables as weight status, race, gender, and income. There were over 750 in each ten-year age group between the ages of 25 and 64 years, and over 200 respondents in each of the other age groups. There were over 3,000 non-Latino White respondents, over 1,000 non-Latino Black respondents, and just over 400 Latino respondents. There were few respondents in other race or ethnic categories, so those are omitted from most analyses in this report that stratify results by race. Household incomes were divided into four groups according to whether the income was below, close to, somewhat higher, or well above the Department of Health and Human Services Federal Poverty guidelines (FPG). The FPG guidelines take into account the number of household residents, when judging household income levels.

The weighted percentages do not reflect the actual percent of respondents in each category. Instead, they represent an estimate of the prevalence of each category among the Marion County adult population, weighted for the sampling scheme used.

Specifically, because a stratified sampling frame was used for this survey, different weights were given to the responses of different respondents, depending on the different likelihood of them being selected for the survey. The final dataset was also weighted to approximate the race, gender, and age distribution of the Marion County adults in general. Consequently, dividing the number of responses shown for a group by the total number of respondents shown will usually not equal the percentages shown. This is true of all the statistics presented in this report.

Table 1: Demographic and Socio-Economic Characteristics of Survey Respondents (n = 4,784)

	Percent[g] (95% CI)[h]		Respondents[i]
Age Group (years)
18 to 24	8.4	(7.3-9.4)	267
25 to 34	24.6	(23.1-26.1)	849
35 to 44	21.9	(20.6-23.3)	858
45 to 54	16.3	(15.3-17.3)	934
55 to 64	13.7	(12.8-14.6)	783
65 to 74	7.9	(7.3-8.6)	550
75 or older	7.2	(6.6-7.8)	502
Gender
Male	48.2	(46.7-49.8)	1,921
Female	51.8	(50.2-53.3)	2,863
Race/Ethnicity
White non-Latino	69.4	(68.1-70.7)	3,194
Black non-Latino	23.8	(22.5-25.1)	1,031
Latino	4.5	(4.2-4.8)	407
Other Race/Ethnicity	2.3	(1.8-2.8)	96
Gender and Race
White Male	32.9	(31.4-34.4)	1,187
Black Male	10.7	(9.7-11.6)	448
Latino Male	2.8	(2.5-3.1)	217
Male, Other or Unknown Race	1.9	(1.5-2.4)	69
White Female	35.7	(34.3-37.1)	2,007
Black Female	12.8	(11.8-13.8)	583
Latino Female	1.7	(1.5-1.9)	190
Female, Other or Unknown Race	1.6	(1.2-1.9)	83
Household Income as a percent of the Federal Poverty guidelines
Less than 100% of the FPG	5.9	(5.1-6.6)	270
100% to less than 200% of FPG	14.4	(13.3-15.5)	670
200% to less than 300% of FPG	17.0	(15.8-18.2)	733
Over 300% of FPG	62.7	(61.2-64.3)	2,546
Education
No high school degree	8.7	(7.9-9.6)	475
High school graduate	28.3	(26.9-29.7)	1,369
Some college	25.7	(24.3-27.0)	1,203
College graduate	37.3	(35.8-38.8)	1,648

Figure 2: Percent of Non-Pregnant Marion County Residents, Age 18 Years or Older, by BMI Category, 2005

Table 2: Prevalence of BMI Categories

BMI Category	Estimated Prevalence	95% Confidence Interval
Underweight	1.4%	1.8%-1.0%
Normal	37.9%	39.5%-36.4%
Overweight	35.1%	36.6%-33.5%
Obese	21.6%	22.9%-20.3%
Morbidly Obese	4.0%	4.6%-3.4%

Among adults in Marion County who were not pregnant, 1 percent had BMI values in the Underweight range (less than 18.5), 38 percent had BMI values in the Normal range (18.5 to less than 25), 35 percent had BMI values in the Overweight range (25 to less than 30), and 26 percent had BMI values in the Obese range (BMI of 30 or more), including 4 percent who were Morbidly Obese (BMI of 40 or more).

NOTE: In the rest of this report, those in the Underweight BMI category were combined with those in the Normal BMI category. Likewise, those in the Morbidly Obese category were combined with those in the Obese BMI category.

The BMI distribution by various demographic categories, and the number of non-missing respondents in each category are shown in Table 3.

Table 3: BMI Distribution by Demographic and Socio-economic Characteristics

	Percent[j] (95% CI[k]) per BMI Category			Number of Respondents[l]
	Underweight or Normal (U-N)	Overweight (Ovr)	Obese (Ob)	U-N	Ovr	Ob
Age Group (years)
18 to 24	59.1 (52.5-65.6)	25.0 (19.3-30.6)	16.0 (10.9-21.0)	149	65	37
25 to 34	47.0 (43.3-50.7)	30.0 (26.6-33.5)	23.0 (19.9-26.1)	373	228	182
35 to 44	36.2 (32.8-39.6)	36.4 (32.9-39.9)	27.4 (24.2-30.6)	311	289	219
45 to 54	32.1 (29.0-35.3)	37.1 (33.8-40.4)	30.8 (27.7-33.8)	298	319	278
55 to 64	29.0 (25.7-32.3)	39.9 (36.3-43.5)	31.1 (27.8-34.5)	222	292	237
65 to 74	35.0 (30.8-39.1)	38.0 (33.7-42.2)	27.1 (23.2-31.0)	183	201	142
75 or older	40.9 (36.4-45.4)	43.1 (38.5-47.6)	16.0 (12.6-19.4)	196	203	76
Gender
Male	33.8 (31.4-36.2)	41.9 (39.5-44.3)	24.3 (22.2-26.4)	585	823	465
Female	44.8 (42.8-46.8)	28.3 (26.6-30.1)	26.9 (25.1-28.7)	1161	785	716
Race/Ethnicity
White non-Latino	41.7 (39.9-43.6)	34.9 (33.1-36.7)	23.3 (21.7-24.9)	1282	1065	734
Black non-Latino	32.0 (28.6-35.5)	33.8 (30.4-37.1)	34.2 (30.9-37.6)	281	354	342
Latino	37.2 (31.8-42.7)	40.7 (35.1-46.3)	22.0 (17.5-26.5)	122	133	79
Other Race/Ethnicity	42.5 (32.0-53.0)	40.2 (29.7-50.7)	17.3 (9.2-25.3)	42	36	16
Gender and Race
White Male	33.9 (31.0-36.7)	42.4 (39.4-45.3)	23.7 (21.2-26.2)	380	511	287
Black Male	33.3 (27.7-38.9)	38.2 (32.8-43.6)	28.4 (23.4-33.5)	126	189	126
Latino Male	32.5 (25.5-39.6)	47.1 (39.6-54.5)	20.4 (14.6-26.2)	57	89	41
Male, Other/Unknown Race	35.7 (23.6-47.9)	48.3 (35.9-60.8)	16.0 (6.9-25.0)	22	34	11
White Female	49.3 (47.0-51.6)	27.8 (25.7-29.9)	22.9 (21.0-24.8)	902	554	447
Black Female	30.9 (26.7-35.0)	29.7 (25.7-33.8)	39.4 (35.1-43.7)	155	165	216
Latino Female	45.8 (37.4-54.2)	29.2 (21.8-36.6)	25.0 (18.0-32.0)	65	44	38
Female, Other/Unknown Race	50.8 (39.1-62.5)	28.8 (18.3-39.3)	20.4 (10.8-29.9)	39	22	15
Household Income as a percent of Federal Poverty guidelines (FPG)
Less than 100% FPG	33.2 (26.3-40.0)	33.3 (27.0-39.6)	33.6 (27.1-40.0)	71	90	84
100% to < 200% FPG	36.7 (32.5-40.8)	31.6 (27.7-35.4)	31.8 (27.8-35.7)	228	216	197
200% to < 300% FPG	39.5 (35.5-43.5)	32.1 (28.3-35.8)	28.4 (24.8-32.1)	272	236	195
Over 300% FPG	39.8 (37.7-41.9)	37.7 (35.6-39.7)	22.5 (20.8-24.3)	974	918	578
Education
No high school degree	38.0 (32.7-43.3)	33.7 (28.8-38.6)	28.3 (23.7-32.9)	148	162	131
High school graduate	37.0 (34.0-39.9)	34.1 (31.3-37.0)	28.9 (26.2-31.6)	467	448	371
Some college	36.0 (32.9-39.0)	36.0 (33.0-39.0)	28.0 (25.2-30.9)	404	425	319
College graduate	43.9 (41.3-46.5)	35.6 (33.1-38.1)	20.5 (18.4-22.6)	705	557	337

Figure 3: Comparison of Overweight and Obesity Between the United States, Indiana, and Marion County

Obesity (BMI over 30) is more prevalent in Indiana and Marion County than in most of the United States. Twenty-seven percent of adults in Marion County and 27 percent in Indiana are obese, compared to a median of 24 percent across the United States.

SOURCE: USA and Indiana statistics: CDC BRFSS 2004 survey. See http://apps.nccd.cdc.gov/brfss/list.asp?cat=DE&yr=2004&qkey=4409&state=All and http://apps.nccd.cdc.gov/brfss/display.asp?yr=2004&cat=DE&qkey=4409&state=IN .

Figure 4: Percent of Adult Females in Each BMI Category, by Race and Ethnicity

Black females had a 70 percent higher prevalence of obesity (39%), than White (23%) or Latino females (23%), and a correspondingly lower prevalence of BMI values in the Normal or Underweight category. In no other demographic group was obesity so clearly the most prevalent BMI category.

In contrast, White and Latino females had a high prevalence of BMI values in the Normal or Underweight range compared to all other demographic groups, including other groups defined by gender, race, age, education, or income.

Note: The race and ethnicity categories were mutually exclusive. All respondents indicating Latino ethnicity are in the Latino category. The White and Black categories do not include respondents indicating Latino ethnicity. The Other Race category is not shown, as it included only 26 female respondents, which were too few for reliable estimates to be calculated.

Figure 5: Percent of Adult Males in Each BMI Category, by Race and Ethnicity

Black males had a higher prevalence of obesity (28%) than did White males (24%). White males had a higher prevalence of obesity than did Latino (19%) males. Almost half of Latino males were overweight.

The differences by race between males were not nearly as great as the differences between females. The BMI distributions of the male racial groups all had roughly the same shape, with overweight being most common, and obesity being least common. In contrast, the BMI distributions among females (Figure 4) clearly differed by race or ethnicity. Female BMI values were skewed in a more healthy direction among White and Latino women, and in a less healthy direction among Black women.

Note: The race and ethnicity categories were mutually exclusive. All respondents indicating Latino ethnicity are in the Latino category. The White and Black categories do not include respondents indicating Latino ethnicity. The Other Race category is not shown, as it included only 23 male respondents, which were too few for reliable estimates to be calculated.

Figure 6: Percent of Adults in Each BMI Category, by Age Group

Marion County adults who were 45 to 64 years old had the highest prevalence of obesity, while the 18 to 24 and 75 and older age groups had the lowest prevalence of obesity. The percentage of Marion County adults who are overweight increases with age.

Figure 7: Percent of Adults in Each BMI Category, by Household Income as a Percent of the Federal Poverty Guidelines (FPG)

There was an inverse relationship between household income and Body Mass Index. As income decreased, the obesity rate increased.

In 2004, 100 % of the Federal Poverty Level for a family of four in Indiana was $18,850 per year.[m] A family of four at or above 300% of the Federal Poverty Level would have had an annual household income of $56,550 or more.

Figure 8: Map of Adult Obesity Prevalence by Health Planning Area within Marion County

The Figure 8 map shows estimates of obese adults (BMI ≥ 30) as a percent of all adults in a health planning area.

There was a moderate amount of variation in obesity rates across Marion County (range : 15.5 to 36.3 percent Obese, by Health Planning Area). The highest prevalence of obesity occurred in a band running from east to west all the way across the county, slightly north of the center of the county. The lowest prevalence was in the Meridian corridor, toward the north border of the county.

Each rate in the Figure 8 map has somewhat wide confidence interval, plus or minus at least six percent. Differences of two or three percentage points are not meaningful.

The Marion County Health Department has developed health- planning areas to aid in reporting health statistics for the county. Each area is defined so as to have a population of 40,000 to 50,000.

[a] Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA. Mar 10 2004;291(10):1238-1245.

[b] Finkelstein, EA, Fiebelkorn, IC, Wang, G. State-level estimates of annual medical expenditures attributable to obesity. Obesity Research 2004;12(1):18–24.

[c] Body Mass Index equals 703 times Weight in pounds divided by (Height in inches²) . BMI values of 18.5 to less than 25 is considered normal, 25 to less than 30 is considered overweight, and 30 and over is considered obese. See Table 6. NOTE: BMI is an indicator, not a definitive measure of body fat. Having a lot of muscle, rather than fat, can cause a high BMI. For instance, body builders have very high BMIs, and would often fall in the Obese BMI range.

[d] Six percent (6%) of Marion County adults were in households with incomes less than the federal poverty level. See Table 1.

[e] Thirty-seven percent (37%) of Marion County adults were in households with incomes less than three times the federal poverty guideline. See Table 1.

[f] BMI equals 703 X Weight (lbs.) / (Height (in.)²). A BMI of 18.5 to 24.9 is considered Normal, 25 to 29.9 is considered Overweight, and 30 and over is considered Obese. See Table 4: Conversion of Height and Weight to Body Mass Index. BMI is an indicator, not a definitive measure of body fat. Highly muscular, lean individuals, for example, body builders, also can have a high BMI, and may fall in the Obese BMI category.

[g] The percents given are weighted percents and are estimates of the prevalence of each category in the population surveyed. The number of respondents in each category does not correspond to the percentages because the survey used a stratified sampling frame. Some subgroups were over-sampled to assure sufficient numbers of responses for subgroup analyses, and responses were re-weighted to reflect the age, race, and gender distribution of the target population

[h] 95% CI: 95 percent confidence interval, reflecting the potential error in the percent estimate.

[i] The number of respondents within a category may not equal the total number of survey respondents, due to responses of “Refuse” or “Don’t know”.

[j] The percents given are weighted percents, and are estimates of the prevalence of each category in the population surveyed. See footnote g.

[k] 95% CI: 95% confidence interval, reflecting the potential error in the percent estimate.

[l] The number of respondents within a category may not equal the total number of survey respondents, due to responses of “Refuse” or “Don’t know”.

[m] http://www.liheap.ncat.org/tables/FY2005/POP05.htm